Is the EU’s investment agreement with China underrated? 

Uri Dadush is a Non- Resident Fellow, and André Sapir a Senior Fellow, at Bruegel The European Union is very open to foreign direct investment. By comparison, despite considerable liberalisation in the past two decades, foreign investors in China’s markets still face significant restrictions, especially in services sectors. Given this imbalance, the EU has long sought to improve the situation for its companies operating or wanting to operate in China. After eight years of negotiations, the EU and China concluded in December 2020 a bilateral Comprehensive Agreement on Investment (CAI). The text awaiting ratification aims to give foreign investors greater market access, enforceable via state-to-state dispute settlement. It does not yet, however, cover investor protection (such as against expropriation). Meanwhile, investor protection is covered by bilateral investment treaties between EU countries and China, which remain in force. The CAI has been met in some quarters with scepticism on economic and geopolitical grounds. The main criticism is that it provides little new market access in China, and that this small economic gain for the EU comes at the price of breaking ranks with its main political ally, the United States. Our assessment, which focuses on the economic implications, is different. It is true the CAI provides only modest new market access in China, but this is because China has already made progress in recent years in liberalising its foreign investment regulations unilaterally. The CAI binds this progress under an international treaty, marking an improvement for EU firms insofar as their market access rights can be effectively enforced. Most important, the CAI includes new rules on subsidies, state-owned enterprises, technology transfer and transparency, which will improve effective market access for EU firms operating in China. These bilateral new rules could also pave the way for reform of the multilateral rules under the World Trade Organization, with the aim of better integrating China into the international trading and investment system – a goal shared by the EU, the United States and other like-minded countries. From an economic viewpoint therefore, the CAI is an important agreement, and one worth having. However, its ratification by the European Parliament is unlikely while China continues to apply sanctions against some members of the European Parliament and other critics of China’s human rights record. The Comprehensive Agreement on Investment Since its beginning, the European Union has maintained a treaty-based policy of openness towards foreign direct investment (FDI), though this varies slightly from EU country to EU country since members retain some prerogatives over FDI. By comparison, China remains restrictive, despite having liberalised its FDI regime in recent decades. Restrictiveness is most notable in the services sector. Investment in China’s manufacturing sector is now quite free, though still less than in the EU. Given the imbalance between the EU and China in terms of investment openness, the EU has long sought to improve the situation for its companies operating or wanting to operate in China. After eight years of negotiations, the EU and China concluded in December 2020 a bilateral Comprehensive Agreement on Investment (CAI)1. Foreign firms wanting to invest in China have always needed to respect the country’s prevailing Foreign Investment Law (FIL), which sets out the general principles applicable to foreign investment. The latest FIL was adopted on 15 March 2019 and entered into force on 1 January 20202. Under the FIL, foreign investment in China in certain industries is explicitly encouraged, while it is prohibited or restricted in others. A typical restriction limits foreign ownership and requires foreign firms to form joint ventures (JV) with local partners. Another typical restriction forbids investment in new capacity in sectors such as steel and cement where overcapacity exists. Although there are many situations in which foreign firms may be treated less favourably than domestic firms, the most recent FIL mandates that Chinese government procurement at all levels of government shall not discriminate between foreign and domestic firms. The FIL also forbids, in principle, forced technology transfer. Importantly, the FIL specifies that when international agreements to which China is a party contain provisions more favourable to the admission of foreign investors, those pro- visions will take precedence over the existing Chinese FDI regulations. International agreements signed by China that include investment provisions include the World Trade Organization agreements and in particular China’s schedule of commitments under the General Agreement for Trade in Services (GATS) schedule; the Regional Comprehensive Economic Partnership (RCEP) agreement between China and 14 other Asia-Pacific countries; and the Phase One US-China agreement. The CAI is the first investment-only liberalisation agreement to which China is a party. The text agreed in December 2020 does not cover investor protection against discrimination, expropriation or denial of justice in the host country. It specifies, however, that China and the EU will complete negotiations on investment protection (and the related investor-to-state dispute settlement) within two years of the signature of the current agreement. In the meantime, bilateral investment treaties (BITs), which cover only investment protection, between individual EU countries and China remain in force. This paper assesses the gains that EU firms planning to invest in China, or that have already done so, will derive from CAI in terms of better market access. We do not mean to overlook the potential economic implications of increased investment in Europe by Chinese firms in the wake of the CAI, but we recognise that, since EU markets are already largely open to Chinese investment, the most likely and largest source of gains for the EU will derive from the new opportunities in China. Another purpose is to evaluate whether and how CAI could help better integrate China into the rules-based multilateral system – an important goal for EU trade and investment policy. We focus on the economic aspect of the agreement and leave aside geopolitical considerations. This does not mean, however, that we do not appreciate strategic issues and how they relate to the workings of the world economy. EU-China relations are, by essence, of systemic importance given that the two parties are the world’s second and third largest economic blocs. At the same time, we are obviously aware that recent clashes between the EU and China over human rights violations in China cast a deep shadow over the agreement. Indeed, ratification of the CAI by the European Parliament is very unlikely while China continues to apply sanctions against some members of the European Parliament and European researchers who have criticised its human rights record. The economic implications of the agreement are themselves the subject of considerable controversy. The CAI has been presented by the European Commission as a major success but has been met with considerable scepticism on the grounds that it does not go far enough3. Our assessment is different. We recognise that the CAI is an imperfect agreement which – in contrast to some of the Commission’s claims – provides only modest new market access in China. However, this is in no small measure due to the progress that China made over the last several years in liberalising its foreign investment regulations unilaterally. The CAI binds this progress under an international treaty, an improvement for EU firms insofar as their market access rights can be effectively enforced. Most important, the CAI includes new rules on subsidies, state-owned enterprises (SOEs), technology transfer and transparency, all of which are of significance for the world trading system and which – with some caveats – increase the likelihood that European investors can operate in China successfully and sustainably. Our overall conclusion is that the CAI is far from a perfect deal, but – from the economic viewpoint – is an important agreement worth having. The impact of CAI on EU FDI in China To what extent does CAI improve effective market access for EU firms already operating in or wishing to operate in China? In principle CAI could improve market access by allowing EU firms to invest in more sectors and/or with less stringent conditions – no longer needing to participate in joint ventures with Chinese firms, for example. Or, the CAI could include rules, such as on subsidies, which will allow EU firms to compete on a more equal footing with Chinese firms in their home market. We look at these two issues in turn. Does CAI improve market access for EU firms in China? Two decades ago, China had a very restrictive investment regime. In 1997, it scored 0.625 on the 0 (open) to 1 (closed) scale of the Organisation for Economic Co-operation and Development’s FDI Regulatory Restrictiveness Index4. By 2019, it scored 0.244, a huge improvement, though China was still the eighth most restrictive out of a sample of 83 countries, including 45 non-OECD members5. Compared to its fellow BRICS, China’s FDI regime is about as restrictive as India’s and Russia’s, but far more than Brazil’s and South Africa’s. It should be borne in mind that the recent revisions to the FIL may not be fully reflected in the 2019 OECD index. For the record, all EU countries have scores near zero (Figure 1 and Table 1).
The CAI demonstrates the EU’s capacity to negotiate successfully and independently with China; the EU has signalled clearly that it has no intention to decouple from China


Figure 1. The OECD’s FDI Regulatory Restrictiveness Index for China, 1997-2019

Note: a score of 0 equals completely liberalised. A score of 1 indicates closed. Source: Bruegel based on OECD.

Table 1. OECD’s FDI Regulatory Restrictiveness Index for China and selected countries, 2019

Note: a score of 0 equals completely liberalised. A score of 1 indicates closed. Source: Bruegel based on OECD. China’s substantial investment liberalisation during the past two decades is attributable to a combination of unilateral and multilateral measures. Consider first China’s manufacturing sector, which is about 60 percent larger than the EU’s and is growing faster, and where most EU FDI is presently directed. According to a 2019 McKinsey study (Woetzel et al, 2019), Chinese manufacturing is already highly integrated into global value chains, using global standards 90 percent of the time, while importing many advanced components. Multinational firms have greater penetration in Chinese consumer markets than they do in the United States (Woetzel et al, 2019). According to UNCTAD, in 2020 China became the world’s largest FDI destination, passing the United States. Shifting the focus from outcomes to regulations, China’s FDI Restrictiveness Index in manufacturing decreased from 0.379 in 1997 to 0.073 in 2019. This means that China’s investment regime in the manufacturing sector is now close to being completely liberalised. For example, according to this OECD statistic, China’s regulations on inward FDI in manufacturing are now more liberal than those of Australia, Canada and Mexico (with scores in the 0.08 to 0.1 range), although not as liberal as Portugal or Morocco, for example, where inward FDI in manufacturing is completely liberalised (score of 0). China’s liberalisation of foreign investment in manufacturing was achieved on a purely unilateral basis, through successive FILs6. By contrast, in the services sector, investment liberalisation has taken place through a combination of unilateral and multilateral measures, since the WTO covers liberalisation of investment in services but not in manufacturing7. Also in contrast to manufacturing, investment liberalisation in services is far less advanced. From 1997, China’s FDI restrictiveness score decreased from 0.739 to 0.306 in 2019 – still much higher than the average score of less than 0.06 in the three biggest EU countries. Like other WTO members, China made multilateral commitments in its GATS schedule under Mode 3 (commercial presence of foreign services companies), when it acceded to the WTO. Among the 162 services sectors listed by the WTO in its Services Sectoral Classification List (SSCL)8, and after a transition period ranging between two and six years after its WTO accession, China made full market access commitments under Mode 3 (meaning that foreign firms can invest freely in China) in only 26 sectors; partial commitments (meaning that foreign firms wishing to invest in China must form a joint venture with a local partner or are subject to other forms of market access limitations) in 71 sectors; and no commitments (meaning that foreign firms are not free to invest in China, unless they obtain specific authorisation) in 65 sectors9. Considering China’s limitations on national treatment, which imply that foreign firms are subject to certain requirements that do not apply to local firms, only 22 of the 162 services sectors are completely free of market access and national treatment limitations under Mode 3 in China’s WTO commitments, a much lower figure than in most other countries’ WTO commitments10. In addition to its multilateral commitments, China has also liberalised some of its services sectors unilaterally through successive FILs. Like in the manufacturing sector, however, liberalisation undertaken unilaterally through national law is, by definition, not bound by a treaty obligation, and can therefore, be revoked at the stroke of a pen. Although important for foreign investors, such unilateral liberalisation is therefore less valuable, because it is less predictable than liberalisation enshrined in an international treaty or agreement. So, what can we say about the CAI? Does it create fresh market access opportunities for EU investors in China compared to the 2019 FIL and China’s WTO commitments? In the manufacturing sector, the CAI binds China’s unilateral liberalisation for the benefit of EU firms. Examination of the commitments in the CAI’s Annex 3 (the positive list) shows that investment in 30 manufacturing sectors is liberalised. Of these, 20 are free of any limitations, including any joint-venture or ownership requirements. These sectors cover a vast array of manufacturing, including food processing, apparel and textiles, chemicals (except for explosives), pharmaceuticals (except for certain types of vaccines), aircraft and spacecraft manufacturing, electrical machinery and equipment, computers and instruments. In the ten sectors for which limitations on market access remain, there are no joint-venture requirements, and the limitations are justified mainly by concerns about overcapacity. For example, in printing, and in petroleum refining China’s schedule in Annex 3 states “increasing production capacity for oil refining shall be in line with the planning.” In sectors including cement, steel, aluminium and shipbuilding, adding production capacity is forbidden for foreign firms as it is for Chinese firms. This means that foreigners can still invest in those sectors, for example by acquiring a Chinese firm or by retooling to adapt their product mix, but they cannot expand overall plant capacity. Examination of CAI Annexes 1 and 2, which contain various exceptions to national treatment, shows no significant provisions on manufacturing, except in the automotive sector. The automotive sector deserves special attention because it is important for EU companies, representing about 30 percent of their total FDI in China. Here there are some access limitations (such as 50 percent Chinese ownership), though these apply only until 2022, in line with the 2019 FIL. There are no joint-venture requirements after 2022. Establishing new production capacity in electric vehicles is allowed but is subject to limitations that apply if there is overcapacity in the designated province. Most important, reflecting China’s intention to promote electric vehicles, China’s schedule in Annex 3 states that “the establishment of new traditional fuel-powered motor vehicle enterprises is prohibited” and increasing capacity in traditional vehicles is subject to model and geographical restrictions related to overcapacity. In the services sector, the CAI needs to be compared to China’s WTO commitments under Mode 3. This comparison shows three main improvements from the CAI. First, China completely opens to foreign investment in eight sectors that were previously closed in its WTO schedule: veterinary services, services related to management consulting, placement and supply services of personnel, telephone answering services, money broking, motor-vehicle financing by non-bank financial institutions, sporting services, and supporting services for rail transport. Second, China partially opens to foreign investment in 11 sectors that were previously closed in its WTO schedule: database services, R&D services for natural sciences, interdisciplinary R&D services, printing and publishing, market research and public opinion polling services, trading of derivative products including futures and options, asset management, hospital services, entertainment services, passenger air transportation services, and freight air transportation services. In these activities, China retains some limitations on market access, such as joint-venture requirements. Third, in most other sectors for which China had previously made only partial commitments, the obligation to form joint ventures has been removed. The main exceptions are some audio-visual services, most telecommunications services and all educational services, where China’s concern seems to be more political than economic11. A detailed comparison of China’s services commitments with those already available to foreign investors under the FIL lies beyond our scope. It is worth noting, however, that in some sectors for which the European Commission claims significant progress on market access, such as telecommunications services, joint-venture requirements remain, and, in the case of hospital services, so do geographical limitations. There is also a complete prohibition on the provision of internet access services. In conclusion, the CAI appears to give only modest new market-access opportunities to EU investors in China compared to the current situation. What it does, however, is to give them the certainty that the big improvements of recent years in market access and national treatment, as reflected in successive FILs, will not be reversed unilaterally by China. The removal of the obligation to form joint ventures in most sectors is of special importance, as it is related to the thorny issue of forced technology transfer, which we discuss in the next section.
Does CAI improve rules and the level playing field for EU firms in China? Being able to access the Chinese market in the framework of an international treaty is obviously important for EU investors, but often they face additional problems in China that the CAI tries to remedy. We focus on four important issues: forced technology transfer, state-owned enterprises, subsidies and standard setting. CAI bans forced technology transfers in covered sectors European and other foreign companies established or seeking to establish in China have long complained that China uses foreign ownership restrictions, including joint-venture requirements, to force them to transfer technology (TT) to Chinese entities. China’s WTO commitments have been of limited help in tackling this. In goods sectors, WTO rules apply to trade but not FDI. Hence, in these sectors, China is free to apply ownership restrictions to foreign companies that may lead to unfavourable TT arrangements for foreign investors. In services sectors, WTO rules apply to FDI, but only to the extent that WTO members have assumed specific obligations under GATS Mode 3. As discussed above, China has assumed some Mode 3 obligations in the schedule of concessions for services attached to its Protocol of Accession to the WTO, but some major sectors are excluded, and in some covered sectors, the establishment of foreign firms is conditional on entering into a joint-venture arrangement with a Chinese entity, which may lead to forced TT. Given the limited ability of current WTO rules and commitments to deal with the problem of forced TT in China, some of its trade and investment partners, primarily the United States, China’s Asian neighbours and the European Union, have sought bilateral or regional solutions. It is important to note that in the 2019 FIL, China responded to these concerns by outlawing forced TT. As in the case of many market-access provisions, however, this falls short of a commitment enforceable under an international treaty. The CAI agreement comes on the heels of efforts to impose disciplines on forced TT on China under the Trans-Pacific Partnership/Comprehensive and Progressive Agreement for Trans-Pacific Partnership (of which China is not a member but which had China in mind), and the China-US Phase One agreement. CAI contains the strongest language yet in that regard. Specifically, CAI contains an obligation for the parties not to “impose or enforce any requirement or enforce any commitment or undertaking … to transfer technology, a production process, or other proprietary knowledge to a natural person or an enterprise in its territory.” As noted by Mavroidis and Sapir (2021), the words ‘impose’ or ‘enforce’ are crucial. They imply that states (the parties to the agreement) cannot impose TT requirements on foreign firms that want to invest in their jurisdictions and that, in case such firms decide to do business in their jurisdictions through a joint venture, the local partner will not be able to enforce any commitment for TT that it may have extracted from the foreign partner as a condition for the joint venture. In principle, therefore, even when joint-venture requirements continue to apply, Chinese companies will no longer be able to force unwanted technology transfer on their EU partners. If enforced properly, this measure, which applies to permitted investment in both goods and services sectors, would be a major advance for EU firms investing in China. CAI makes headway in dealing with the problem of unfair competition from (SOEs) For the first time in an international agreement to which China is a party, CAI contains a precise and comprehensive definition of SOEs, by applying rules to SOEs at all levels of government, including local government, and by improving transparency. State-owned enterprises are omnipresent in the global economy and the WTO agreement places no restriction on their operation, provided they operate on a commercial basis. However, well before China’s WTO accession, and increasingly since, concerns about the competitive distortions caused by China’s large and opaque SOE sector have been prevalent. At the time of China’s WTO accession, SOEs accounted for a large share of economic activity in China They still account for between 23 percent and 28 percent of GDP in China today (Zhang, 2019), compared to about 15 percent in the EU. Surprisingly, given these concerns, China’s accession protocol to the WTO (henceforth ‘the protocol’) includes only a few paragraphs that relate to SOEs. The main thrust of China’s commitments were: a) SOEs would buy and sell only based on commercial considerations and the government would not directly or indirectly influence SOEs’ commercial decisions; b) any subsidy to a SOE would be considered specific, hence actionable; c) SOEs would be responsible for their own profits and losses; d) China will notify any subsidy given to an SOE. A notable omission from the protocol was a definition of an SOE, which proved a cause of confusion and major disputes. The CAI refers to SOEs as ‘Covered Entities’, and establishes criteria to recognise them. These criteria go beyond full or majority ownership to include the power of the state to appoint directors and to control the decisions of the enterprise through “any other ownership interest” or even without “any ownership stakes.” Firms granted monopolies by the state are also defined as SOEs. The definition applies at “all levels of government”, which includes the operation of SOEs owned by local and regional government. Under Section 2 Article 3 of the CAI, covered entities must “act in accordance with commercial considerations in the purchases and sales of goods or services in the territory of the Party …” and not discriminate. The CAI does not include specific provisions on notification of subsidies to SOEs (or by SOEs). Instead, subsidies to SOEs are covered under the new notification requirement (services) and under ‘Consultations’ in the Article on Transparency of Subsidies (for goods and services), which apply to all enterprises. However, the CAI establishes a procedure through which a party can demand information about a subsidy, or on an entity that is believed to be subsidising, where it believes its interests are being adversely affected. The information that the requested party must provide is detailed in CAI Section 2 Article 4, and it includes information on ownership, the entity’s revenue, the size of the subsidy and its purpose. Compared to recent trade agreements involving China, the CAI section on covered entities is an improvement. Neither the China-US Phase 1 Agreement nor RCEP includes disciplines on SOEs. China was not a party to the negotiations of the US-led and subsequently abandoned Trans-Pacific Partnership (TPP). However, Chapter 17 of that agreement covers SOEs and was designed partly with China in mind. It is sometimes referred to as the gold standard on SOE disciplines. Thus, it is no mean achievement that the scope of SOE commitments under the CAI is similar and compares quite favourably with those of the TPP’s successor agreement, the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP). In fact, whereas the TPP’s provisions apply only to central government, the CAI’s apply to all levels of government, a significant improvement given the importance of China’s province- and local government-owned enterprises. Moreover, the CAI definition of SOEs is more encompassing than that of the TPP, including SOEs for which the government can control decision-making. However, the TPP demands more transparency than the CAI, since it obligates the parties to provide a list of all SOEs within two years of ratification of the agreement. CAI provides new disciplines on subsidies in the covered service sectors The CAI’s disciplines on subsidies in services represent a major improvement since the WTO only covers subsidies in goods trade. Overseas investors in China have long complained of a lack of transparency and unfair competition from subsidised Chinese entities. Subsidy disciplines under the Agreement on Subsidies and Countervailing Measures (SCM) and China’s accession protocol date back to when China was a relative minnow in global trade. As China rose to be the world’s largest trading nation, dissatisfaction with aspects of these agreements grew. The dissatisfaction is heightened by the increasing importance of services in global economic activity and trade, since WTO disciplines do not cover subsidies in services, only goods. The intention to negotiate disciplines on subsidies in services was included in Article XV of GATS, but these negotiations never took off. The most important innovation of CAI is to cover subsidies in the eligible services sectors covered by the CAI. However, the CAI’s enforceable provisions relate only to transparency (notification) and consultation relating to subsidies. If a subsidy above a certain size is found to exist, the subsidising party is not required to remove the subsidy or to accept the complainant’s countervailing measures, only to use its best endeavours to find a solution. The relevant paragraphs in section 3 (Regulatory Framework), Article 8 (Transparency of Subsidies) read: Paragraph 7. If the requesting Party, after the consultations have been held, considers that the subsidy concerned has or could have a significant negative effect on the requesting Party’s investment interests under this Agreement, the requested Party shall use its best endeavours to find a solution with the requesting Party. Any solution must be considered feasible and acceptable by both Parties. And, Paragraph 10. Paragraph 7 shall not be subject to Section X (State to State Dispute Settlement). Neither the China-US Phase 1 nor RCEP include new provisions on subsidies, so the CAI is a step forward. It is also instructive to compare the CAI to the WTO subsidy reforms proposed in January 2020 by the EU as a member of the Trilateral Group, which includes Japan and the United States. The Trilateral Group does not mention China specifically, but the group’s proposals have China very much in mind12. The Trilateral Group proposals relate only to industrial subsidies, not services, but most of the proposals could be directly applied in services as well. In addition to calling for a clear definition of SOEs, which the CAI provides, the Trilateral Group proposed a set of far-reaching subsidy reforms, none of which are envisaged in the CAI, at least explicitly. For example, the Trilateral Group proposals would extend the list of prohibited subsidies to cover state guarantees. The Trilateral Group also proposed that subsidies that are not notified are deemed illegal if they are counter-notified and no written explanation is forthcoming. The implicit reference to China is clearest in the Trilateral Group’s proposal to change the way subsidies are calculated, namely, to replace the use of domestic prices with appropriate benchmarks when the market of the subsidising member is distorted. In conclusion, while the CAI includes important steps forward in subsidy disciplines, deeper reforms, such as those envisaged by the EU and its partners in the Trilateral Group, remain a work-in-progress. CAI helps EU firms participate in Chinese standard setting in covered sectors Technical regulations and standards are important features of modern economic life, which can constitute barriers to international trade and investment. An important condition for foreign firms to have effective access to domestic markets is therefore participation in standard-setting bodies. In the European Union, as elsewhere in the OECD, foreign companies have the same right to participate in standard-setting bodies as EU firms. The situation is different in China. According to a report submitted in 2018 by the US authorities to the WTO13, US companies and other foreign companies operating in China face two types of problems with respect to standard setting. First, Chinese government officials in some cases reportedly have pressured foreign companies that want to participate in the standard-setting process to license their technology or intellectual property on unfavourable terms. Second, China has continued to pursue unique national standards in several high-technology areas where international standards already exist, such as in telecommunications, wireless networks and information security. So far, dealing with this problem has proved difficult. China’s protocol of accession to the WTO contains no obligation on China regarding its standard-setting process. Nor does the US-China Phase One trade agreement. Only the RCEP agreement mentions standard setting, but simply to encourage the exchange of information between the relevant national bodies. The CAI is clearly innovative here. It commits the two parties to allow foreign enterprises that are covered by the agreement “to participate in the development of standards by [their] central government bodies, including related standardisation working groups and technical committees at all levels, on terms no less favourable than those it accords to its own enterprises, including its covered entities.” This commitment is much more valuable to EU firms operating in China than to Chinese firms operating in the EU, since the former currently face much more difficulty than the latter in participating in standard-setting bodies. But it is also a guarantee for Chinese firms that they will continue to be treated like EU firms by government bodies in the EU. It should be noted, however, that the obligation to treat foreign firms no less favourably than domestic enterprises only applies to government standard-setting bodies. In the case of local and non-governmental bodies, the parties have only an obligation to recommend that they provide such treatment. Enforcement Obviously, commitments are only worthwhile if they can be enforced. The CAI’s market-access and level playing field commitments are enforceable via a state-to-state dispute resolution mechanism, as in the EU’s trade agreements with other nations. In case of a dispute, and if the parties fail to reach an agreement through consultations, the complaining party may request the establishment of an arbitration panel. If the panel rules in favour of the complainant and the respondent fails to abide by the decision of the panel and remove the disputed measure within a reasonable period of time, the complaining party may retaliate by adopting an equivalent measure. No international treaty is fool proof, of course, and all treaties depend on the willingness and ability of contracting parties – sovereign nations – to abide by their terms. China’s record on compliance with its trade treaty obligations, notably in the WTO, is no better or worse than that of other nations. It is worth noting, however, that China has been the subject of many WTO disputes and it has a strong record of compliance with decisions of panels and the Appellate Body when they ruled against it. The impact of CAI on the global trading and investment system China is now the world’s largest trading nation in goods, and among the fastest growing. The main systemic implication of the CAI is that it strengthens disciplines on China’s idiosyncratic economic system in important areas. As the European Commission has claimed, the CAI is sui generis, an entirely new type of bilateral investment agreement which covers market access in both the goods and services sectors. Previous bilateral investment agreements – of which there are currently over 2000 in force – cover only investor protection (for example, against expropriation and prohibitions on profit repatriations), not market access. As mentioned, market access through foreign investment in the goods sector is not covered under the WTO, and nor is it covered in China’s accession protocol. In contrast, market access through foreign investment in the services sector is governed under GATS Mode 3 (foreign establishment), negotiated in the Uruguay Round. As discussed in the previous section, China made modest services commitments under Mode 3 in its Protocol, which the CAI deepens and extends. Thus, a notable systemic implication of the CAI is that it may pave the way for new types of bilateral agreements that make international investment in goods and services more open and predictable. This may include progress in the stalled investment negotiations between the US and China. The CAI covers many, but not all, services sectors, and excludes treatment of trade under Modes 1 (cross-border), 2 (consumption abroad) and 4 (presence of natural persons). Therefore, it does not conform to the conditions allowing for an exception to the non-discrimination principle, envisaged for regional trade agreements under GATS Article VI (parallel to GATT Article XXIV). It follows that China’s and the EU’s commitments in services must be extended to all WTO members, under the MFN provision (GATS Article II). Thus, another notable systemic implication of the CAI is that it not only binds and enhances liberalisation of foreign establishment in two of the world’s largest economies, but this is to the benefit of all WTO members, not just the EU. However, since investment in goods is not covered by the WTO, CAI provisions on market access in the goods sector are not MFN, and only benefit the EU. From a WTO perspective, the CAI has some negative and some positive features. The negative feature is that the CAI, because of its bilateral rather than multilateral nature, favours EU over other foreign investors in China, since some of its provisions only apply to EU firms. The most important positive feature is that it advances disciplines in transparency, subsidies, technology transfer and SOEs among two of the world’s largest economies, disciplines which could potentially be incorporated at some stage in the WTO, based on plurilateral or multilateral negotiations involving other WTO members besides the EU and China. Also potentially important for the WTO are the CAI’s sustainable investment provisions, which cover labour and environmental standards. Such provisions are commonly included in the EU’s and the US’s bilateral or regional agreements. However, this is the first time that they have been adopted by China, which has been among the leaders of the developing country resistance against including these disciplines in WTO agreements. Other systemic implications of CAI can be seen from the perspectives of the world’s three largest economies: China, the United States, and the EU. China By adopting the CAI, China has shown that it is willing to consolidate (bind) its progress on investment liberalisation under its FILs into an international treaty, and that it intends to continue to confront Chinese firms with world-class competition on their own turf. In the covered services sectors, the liberalisation is far from complete but affects all foreign investors, not just those from the EU. In goods sectors, the liberalisation has been far-reaching but is internationally bound only as it concerns European firms. However, since these include many of the world’s largest multinationals, the competitive effects on Chinese goods markets could be far-reaching. Moreover, by agreeing to disciplines on transparency about subsidies, SOEs and forced technology transfer, China has taken another step towards promoting ‘competitive neutrality’ between firms that are state-owned (or state-influenced) and private firms, including both Chinese and foreign firms14. Considering the commitments that China made on technology transfer and intellectual property protection in its Phase One deal with the United States, with the CAI China is also signalling, both at home and abroad, that it wants to deal with the major concerns raised by foreign firms competing in China or with Chinese firms on overseas markets. Although China gains little in new market access in the EU (in the market for renewable energy equipment, for example) it scores other important wins. At a time of great trade tensions, the EU binds its commitment to receive Chinese foreign investment with very few limitations. This would not, however, prevent the EU from restricting market access to a Chinese company in case an arbitration panel found that it received state subsidies or acted not in accordance with commercial considerations. In combination with the conclusion of the RCEP negotiations with Asia-Pacific partners, China is showing through the CAI that it is becoming more integrated in global markets and will not be isolated even in the face of potential continued US hostility. China’s unilateral liberalisation and level playing field reforms, consolidated in the CAI, send an important signal to foreign investors that they are welcome, and that China sees their contribution as important for its continued development. The United States There is an assumption15 that the CAI pre-empts EU cooperation with the Biden administration in striking a tougher bargain with China. But it is far from certain, considering thorny China-US relations, that such a bargain was there to be had. As it happens, by agreeing to a temporary suspension of tariffs under the Airbus-Boeing dispute, and in various other ways, the Biden administration has already signalled that it is keen on intensifying its collaboration with the EU on trade matters. More importantly, the CAI clearly furthers US economic interests. The US benefits directly from China’s services liberalisation under the CAI, including the elimination of joint-venture requirements in many sectors. Even though the CAI’s many provisions on transparency, including in subsidies, the definition of what constitutes an SOE, and the strong ban on forced technology transfer apply only to the sectors covered and only to the EU, they directly or indirectly address some of the United States’ most important demands on China. The Biden administration intends to hold China to its Phase One commitments, and – when and if it decides to embark on Phase 2 negotiations – will have the opportunity to coordinate with the EU its position on the outstanding level playing field issues. The EU The EU’s new commitments under the CAI are limited, since its market is largely open and disciplines on subsidies, for example, are already strict. Moreover, as the European Commission has stressed, the EU retains its instruments to screen inward investment from China and elsewhere for any violation of its competitive neutrality criteria. The CAI does, however, demonstrate the EU’s capacity to negotiate successfully and independently with China. In concluding the CAI negotiations, the EU has signalled clearly that it has no intention to decouple from China, a course that some radical factions in the United States and elsewhere advocate. The CAI is both innovative and supportive of the world trading system, in a manner that the highly discriminatory ‘managed trade’ Phase One deal was clearly not. This result was achieved even though the EU had little to give that is new to China beyond binding its investment liberalisation. Since the entry into force of the Lisbon Treaty in December 2009, EU trade and investment policy must ‘be guided’ by fundamental EU principles, including human rights, labour rights and environmental protection. The CAI’s sustainable investment clauses are an important victory for the EU and a major concession by China since – as mentioned – none of the trade agreements it has signed so far, including the RCEP, contains a sustainability clause. The SI chapter legally obliges the parties to the CAI to “effectively implement the ILO Conventions it has ratified and …[to] make continued and sustained efforts on its own initiative to pursue ratification of the fundamental ILO Conventions No 29 and 105 [on forced labour], if it has not yet ratified them.” In case of disagreement between the parties regarding the fulfilment of their sustainable-investment obligations (which cover not only labour, but also corporate social responsibility, the environment and climate change) the matter will be referred to an independent expert panel. The sanction for not respecting sustainable-investment obligations is not retaliation like with market access commitments, but naming and shaming through the publication of the expert panel’s report. This may help China enforce its sustainable-investment obligations, to the extent that it cares about its international image and standing.
Conclusion The CAI is far from a perfect deal, but from an economic viewpoint, it is an important agreement, and one worth having. The CAI falls short of the high expectations set at the time the negotiations were launched in 2014. It is a work in progress, and its provisions relating to investment protection, which were intended to replace the BITs between China and EU member states, remain to be negotiated over the next two years. In contrast to some of the Commission’s claims, the CAI makes only modest advances in new market access relative to that established under China’s succession of FILs. However, it consolidates that level of access in an international treaty. Under the CAI, EU firms will have predictable and, in principle, largely unfettered access for investment in goods and more predictable access in services, where many impediments remain. Joint-venture requirements are eliminated in goods and in many service sectors, thus largely reducing the possibility of forced technology transfers. Also, new rules on subsidies, SOEs and transparency – though still incomplete – level the playing field for EU firms and improve their ability to achieve sustained profitability in China. As the first liberalisation investment-only agreement, the CAI is innovative and of considerable systemic significance, as it tightens the disciplines on China’s socialist market economy and improves the chances that it will become better integrated into the world trading system. The CAI opens new avenues for improving WTO rules on SOEs and subsidies, and for negotiating bilateral investment agreements that include market access. The CAI addresses fundamental concerns about Chinese behaviour shared with the United States, the EU’s most important ally, and directly promotes the United States’ economic interests by consolidating its market access in China’s services sector. If divisions over human rights and geopolitical competition can be bridged, both the EU and China will derive considerable economic benefits from the agreement, and the prospects for sustaining an open and predictable world trading system will improve. Endnotes 1. The text of the agreement and its annexes can be found here: 2. For an extensive discussion of the 2019 FIL, see World Bank (2020). 3. Some of the sceptics are Beattie (2021), García-Herrero (2021), Godement (2021) and Gros (2021). 4. According to the OECD’s website, the FDI Index measures statutory restrictions on foreign direct investment in 22 economic sectors. It gauges the restrictiveness of a country’s FDI rules by looking at the four main types of restrictions on FDI: foreign equity limitations, discriminatory screening or approval mechanisms, restrictions on the employment of foreigners as key personnel, and other operational restrictions, such as restrictions on branching and on capital repatriation or on land ownership by foreign-owned enterprises. Restrictions are evaluated on a 0 (open) to 1 (closed) scale. The overall restrictiveness index is the average of sectoral scores. 5. The seven countries with a higher FDI index than China are: Libya (0.713), Algeria (0.587), Philippines (0.374), Indonesia (0.345), Thailand (0.268), Russia (0.261) and Malaysia (0.252). 6. The only manufacturing sector for which China made a multilateral commitment at the time of its entry to the WTO in 2001 was the automotive sector, with China granting limited market access. Foreign firms still needed authorisation to invest in the automotive sector in China and were required to enter a joint-venture arrangement with a local partner, but they were given freedom to choose the categories, types and models they wanted to produce. 7. A partial exception is the Agreement on Trade Related Investment Measures (TRIMS) which forbids setting conditions on inward FDI such as local content or export requirements. The TRIMS agreement covers only goods. 8. See WTO Document MTN.GNS/W/120/ of 10 July 1991. 9. See Protocol of Accession of the People’s Republic of China (WT/L/432), and the People’s Republic of China Schedule of Specific Commitments (GATS/SC/135). 10. In terms of broad economic sectors, the 22 fully liberalised sectors under Mode 3 in China’s accession protocol include: 9 business services, 6 transport services, 3 distribution services, 2 financial services, and 2 tourism and travel services. At the other extreme of the spectrum, the 65 sectors with no commitments under Mode 3 in China’s GATS schedules include 23 transport services, 15 business services and 9 financial services; there are also no WTO commitments under Mode 3 by China in health services, and recreational services. 11. China will also continue to apply joint-venture requirements in three financial services sectors and three transport services sectors. 12. The EU’s announcement of the Trilateral Group proposals is available at: 13. See WTO Document WT/GC/W/746 of 16 July 2018. 14. See García-Herrero and Ng (2021) for a discussion of SOEs and competitive neutrality in China. 15. Including on the part of those who view the deal sceptically (see footnote 3); see also, for example: References Beattie, A (2021) ‘EU’s investment deal will give it limited inroads into China’, Financial Times, 18 March García-Herrero, A (2021) ‘Europe’s disappointing investment deal with China - Why rush a deal that is so inherently complex?’ Bruegel Blog, 4 January García-Herrero, A and G Ng (2021) ‘China’s state-owned enterprises and competitive neutrality’, Policy Contribution 05/2021, Bruegel Godement, F (2021) ‘Wins and Losses in the EU-China Investment Agreement (CAI)’, Policy Paper, Institut Montaigne, January Gros, D (2021) ‘The Limits of the EU China Investment Agreement’, Project Syndicate, 4 February Mavroidis, PC and A Sapir (2021) China and the WTO: Why Multilateralism Still Matters, Princeton University Press Woetzel, J, J Seong, N Leung, J Ngai, J Manyika, A Madgavkar, S Lund and A Mironenko (2019) China and the World: Inside the dynamics of a changing relationship, McKinsey Global Institute, July World Bank (2020) 2019 Investment Policy and Regulatory Review – China, World Bank, Washington DC Zhang, C (2019) ‘How Much Do State-Owned Enterprises Contribute to China’s GDP and Employment?’, mimeo, World Bank, Washington DC The authors are grateful to Bruegel colleagues and to Petros Mavroidis for helpful comments, and to Mia Hoffmann for superb research assistance. This article is based on Policy Contribution Issue no09/21 | April 2021